SEARCH

SEARCH BY CITATION

Keywords:

  • Antibody-mediated rejection;
  • kidney biopsies;
  • microarrays;
  • rejection;
  • T cells;
  • transplants

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References

Microarray studies of kidney transplant biopsies provide an opportunity to define the molecular phenotype. To facilitate this process, we used experimental systems to annotate transcripts as members of pathogenesis-based transcript sets (PBTs) representing biological processes in injured or diseased tissue. Applying this annotation to microarray results revealed that changes in single molecules and PBTs reflected a large-scale coordinate disturbance, stereotyped across various diseases and injuries, without absolute specificity of individual molecules or PBTs for rejection. Nevertheless, expression of molecules and PBTs was quantitatively specific: IFNG effects for rejection; T cell and macrophage transcripts for T cell-mediated rejection; endothelial and NK transcripts for antibody-mediated rejection. Various diseases and injuries induced the same injury–repair response, undetectable by histopathology, involving epithelium, stroma and endothelium, with increased expression of developmental, cell cycle and apoptosis genes and decreased expression of differentiated epithelial features. Transcripts reflecting this injury–repair response were the best correlates of functional disturbance and risk of future graft loss. Late biopsies with atrophy-fibrosis, reflecting their cumulative burden of injury, displayed more transcripts for B cells, plasma cells and mast cells. Thus the molecular phenotype is best described in terms of three elements: specific diseases, including rejection; the injury–repair response and the cumulative burden of injury.


Abbreviations: 
BFC

biopsies for clinical indication

TCMR

T cell-mediated rejection

ABMR

antibody-mediated rejection

ATN

acute tubular necrosis

PBTs

pathogenesis-based transcript sets

SLC

solute carrier

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References

Molecular assessment of biopsies represents a new dimension in our understanding of disease phenotypes, providing mechanistic insights, potential diagnostic tests, targets for drug discovery and an opportunity to validate and refine histopathology-based diagnostic classifications. Although protein expression is important, detection methods do not permit the study of most proteins in biopsies. Fortunately, mRNA offers unlimited amplification, providing opportunities to study low abundance molecules. While early studies of single transcripts by RT-PCR were valuable in understanding rejection (1), the real opportunity came with the emergence of microarrays (2), permitting genome wide assessment (reviewed in (3)) (4,5).

With the support of a Genome Canada grant, we developed a center to apply microarrays (mostly Affymetrix) systematically to thousands of experimental and clinical specimens and to compare the molecular phenotype to histopathology. What emerged is a comprehensive picture that provides a general template for other molecular studies. This overview is a companion paper to ‘An integrated view of molecular changes, histopathology and outcomes in kidney transplants’ (6) and summarizes our analysis of the organization of transcript changes in transplant biopsies, and their relationship to other findings, for example biopsy lesions, HLA antibodies. (Space limitations preclude a true review of all published microarray papers, but we are also preparing a separate minireview integrating our findings with previous molecular studies.)

Microarray analysis of biopsies

In general, sufficient RNA for microarray analysis can be isolated from one 18 gauge needle biopsy core that has been properly stabilized in RNAlater® to prevent mRNA degradation, consistently yielding 2–5 μg of RNA. Because reproducibility between cores is good, only one core is needed. The presence of some exogenous tissue such as muscle is readily detected and does not interfere with the assessment of the genes and gene sets of interest. (Small fractions of a core can be used but sampling error becomes a concern.) Comparison of transcript expression by microarrays agrees well with RT-PCR results (7): the dynamic range of microarrays, albeit less than RT-PCR, is nevertheless sufficient to detect pathologic changes. Moreover, while RT-PCR is useful and often essential for studying specific molecules with low abundance, there is no need to confirm positive microarray results with RT-PCR (8) when the microarrays are from established commercial suppliers and rigorous controls and quality assessments are included.

Analysis of microarray data

Microarrays produce high dimensionality data, that is many measurements per sample, which can be statistically challenging (9). It is easy to introduce bias and subtle data-mining, and care must be taken by authors and editors to avoid publishing erroneous results (10,11). Relatively straightforward approaches to analysis have emerged, which can be represented transparently. Readers should be skeptical of claims based on analyses that cannot be explained clearly: remember ‘The Emperor's New Clothes’. Molecules can be analyzed singly or as gene sets, and related to other features of the biopsy (lesions, diagnoses) and the patient (clinical findings, function and outcomes).

The objectives of analysis include:

  • (i) 
    class discovery: identifying patterns and groupings in the data;
  • (ii) 
    class comparison: defining differences between groups of samples with predetermined labels, for example antibody-mediated rejection (ABMR) versus T cell-mediated rejection (TCMR);
  • (iii) 
    class prediction: assigning a new biopsy to a class based on its similarity to other biopsies in the database. Class prediction uses weighted equations—‘classifiers’—to estimate the probability of class membership.
  • (iv) 
    Gene set and pathway analysis.

Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References

In ancient Rome, Celsus catalogued changes in inflammation that were stereotyped across many diseases and injuries—rubor, calor, dolor, tumor. Functio laesa was added later. We reasoned that the molecular changes in diseases must also reflect large-scale biologic processes that are stereotyped across different diseases and injuries. Such events are readily apparent, both in mouse and human: entry of T cells and macrophages, IFNG effects and changes in the epithelium, stroma and microcirculation. Using experimental systems (mouse transplants, human cell lines), and in some cases using the literature, we annotated the transcript changes corresponding with these events as pathogenesis-based transcript sets (PBTs) (Table 1) (Figure 1) (http://transplants.med.ualberta.ca/).

Table 1.  Pathogenesis-based transcript sets (PBTs): a system for deconstructing complex changes into biologic mechanisms and interpreting changes in individual transcripts
PBTPBT NameBiological Description
QCATQuantitative cytotoxic T cell-associated transcriptsBurden of effector/effector-memory T cells
GRITInterferon-gamma- and rejection- induced transcriptsInterferon-gamma effects on the tissue and inflammatory cells
QCMATQuantitative constitutive macrophage-associated transcriptsBurden of macrophages
AMAT1Alternative macrophage activation-associated transcriptsAlternatively activated macrophages
ENDATEndothelium-associated transcriptsMicrocirculation response to injury
IRITD3Injury- and repair-induced transcripts day 3Active injury–repair response: ‘injury-up’ Increased in isografts, peaking day 3
IRITD5Injury- and repair-induced transcripts day 5Active injury–repair response: ‘injury-up’ Increased in isografts peaking day 5
KT1Kidney transcripts—set 1Active injury–repair response: ‘injury-down’ Parenchymal transcripts
KT2Kidney transcripts—set 2Active injury–repair response: ‘injury-down’ Solute carrier transcripts
image

Figure 1. Time course of expression of pathogenesis-based transcript sets in isografts (A) and allografts (b) reveals both inflammation and the active injury–repair response. PBTs reflecting inflammation include IFNG effects (GRITs), T cell burden (QCATs), macrophage burden (QCMATs and AMATs). PBTs reflecting the injury–repair response include injury-up transcripts (IRITD3, IRITD5); endothelial cell transcripts, reflecting microcirculation changes (ENDATs); and KTs reduced after injury (KT1, KT2). These changes were assessed in (A) isografts (CBA into CBA); B: CBA naive kidneys with ischaemic ATN assessed 7 days after vascular clamping for 60 minutes; C: allografts (CBA into B6) at days 1–21 posttransplant using Affymetrix microarrays. Expression of each gene set is summarized as the geometric mean for each experimental group (usually three arrays each representing three kidneys) at each time point. PBTs were defined previously (Table 1)(http://transplants.med.ualberta.ca/Nephlab/data/gene_lists.html).

Download figure to PowerPoint

IFNG effects

IFNG- and rejection-induced transcripts (GRITs) were defined in mice as inducible by IFNG, increased in rejecting allografts, and decreased in IFNG-deficient allografts (12). GRITs were transiently increased in mouse isografts, peaking at day 4–5, and in ischemic ATN, as part of the secondary inflammation induced by the active injury–repair response (12,13). In rejecting mouse allografts, expression of GRITs was detectable from day 1, peaking about day 7, then sustained for at least six weeks. GRITs were induced in donor tissue, host macrophages in the graft, and remote host tissues via IFNG released systemically from the graft.

T cell burden

Cytotoxic T lymphocyte (CTL)-associated transcripts that performed quantitatively (QCATs) were defined by their expression in CTL and absence in normal kidney (14,15). There was a small increase in QCATs in isografts and ATN, probably reflecting effector-memory T cells entering the damaged tissue. In allografts, QCAT expression closely paralleled IFNG effects, appearing at day 1, increasing until day 7, then plateauing, indicating that the capacity of the tissue for T cell infiltration was finite and saturable. Maximum QCAT expression in rejecting kidneys was about 15% of that in pure CTL, simulating a sevenfold dilution of CTL in kidney tissue. The T cell burden was established before tubulitis lesions appeared.

We compared transcript expression in human CD4+ CTL, CD8+ CTL and NK cells (16). Transcripts such as GZMB and IFNG were not specific for CTL but were also expressed in resting effector-memory T cells, and were similar between CD8+ and CD4+ CTL. There were no transcripts specific for CD4+ versus CD8+ CTL other than the CD4 and CD8 transcripts themselves. Transcripts were extensively shared between CTL and NK cells, for example NK receptors KLRC1 and KLRD1.

These results have implications for the interpretation of T cell transcripts in tissues, which must acknowledge the similarity between CTL and effector-memory T cells, between CD4+ and CD8+ CTL, and between CTL and NK cells. Thus transcripts such as granzyme B are not specific for TCMR, but also reflect homing of effector-memory T cells to injured tissues, or NK cells in capillaries in ABMR.

Macrophage burden and alternative macrophage activation

Constitutive macrophage-associated transcripts that performed quantitatively (QCMATs) were defined by expression in cultured human macrophages, lack of IFNG inducibility, and low expression in other cell types. Alternative macrophage activation-associated transcripts (AMATs) were identified from the literature (17) and separated from the QCMATs. AMATs are prominent in rejection, especially TCMR, despite strong effects of IFNG, and were also found in injured tissues, reflecting their association with tissue remodeling (17,18).

Injury- and repair-induced transcripts (IRITs): ‘Injury-up’ transcripts

IRITs were defined to reflect the injury–repair response of parenchymal and stromal cells, using mouse kidney isografts and native kidneys with ischemic acute tubular necrosis (ATN) (19). We removed transcripts expressed in inflammatory cells and endothelial cells. Histologically normal isografts displayed hundreds of increased IRITs, which were heterogeneous in their time course. Some peaked at day 1 then declined immediately, and were also increased in the host tissues. Thus some transcripts induced in the transplant at day 1 reflect the systemic acute phase response to wounding/anesthesia rather than simply direct injury.

In contrast, the IRITs that increased more slowly in isografts, peaking at day 3 (IRITD3) or day 5 (IRITD5), reflected the response of the parenchyma, stroma, and microcirculation to injury—the active injury–repair response. The active injury–repair response was largely undetectable by histopathology, and occurred in native kidneys with ischemic ATN as well as in isografts. Many IRITD3 and IRITD5 are known to be TGFB1—regulated, and some recapitulate embryonic development or reflect cell cycle and apoptosis. The response in isografts included expression of many collagen transcripts but did not result in accumulation of fibrosis. Thus collagen gene expression reflects active injury–repair, not scarring. (Fibrosis is linked to tubular atrophy and nephron loss, and results when injury exceeds the ability of the injury–repair response to restore integrity.)

In allografts, the increase in IRITD3 and IRITD5 was similar to that in isografts for 48 hours, reflecting implantation injury, but increased more in allografts after day 3 as the interstitial infiltrate of TCMR appeared. This allospecific augmentation of the IRITs is the essence of TCMR: the interstitial inflammation organized by antigen-specific T cells induces the injury–repair response (19). The pattern of injury-up and injury-down changes in TCMR resembled that in ischemic ATN in native kidneys (20). This is not because TCMR causes ischemia, but because many types of injury and disease evoke the same active injury–repair response.

Induced endothelial cell transcripts reflect microcirculation changes

Many endothelium-associated transcripts (ENDATs) (21) behave as injury-up transcripts, and correlate with IRITD3 and particularly IRITD5. ENDATs primarily reflect the microcirculation, which constitutes the majority of endothelial cells in biopsies. Thus the microcirculation participates in the injury–repair response.

Parenchymal transcripts reduced after injury: ‘injury-down’ transcripts

Kidney transcripts (KTs) were defined to interrogate function and metabolism in parenchymal cells (22), and are defined by high expression in normal kidney. Solute carriers were assigned to a separate set (KT2s) because of their established function and localization. KT1 and KT2 expression was reduced in isografts, ATN, and allografts. KTs slowly recovered in isografts after day 5, but remained low in allografts as TCMR induced the active injury–repair response. Loss of KTs in allografts may explain why injury causes loss of function, that is the molecular equivalent of functio laesa. The injury–repair response (both injury-up and injury-down changes) in TCMR required T cells but was independent of T cell cytotoxic mechanisms, B cells and antibody (19,22).

Summarizing the PBT Changes in Mouse Kidney Transplants

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References

At day 1, kidney isografts showed the early phase of the response to donation-implantation stresses (Figure 1A). By day 2–3, host macrophages began to infiltrate the interstitium of both isografts and allografts. The isografts displayed a mild but complex active injury–repair response, with inflammatory features (QCATs, GRITs, QCMATs), injury-up features (IRITD3, IRITD5), and loss of epithelial features (KT1, KT2). Similar changes were observed in day 7 ATN (Figure 1B). The active injury–repair response in isografts peaked around day 5 then improved but had not completely resolved by day 21. Note that the active injury–repair response is accompanied by secondary inflammation.

Allografts (Figure 1C) display all the features of isografts but even at day 1 also showed specific expression of IFNG effects and T cell transcripts, coinciding with T cell infiltration in perivascular areas (15). Allografts after day 3 displayed progressively more T cell and macrophage transcripts and augmentation of the active injury–repair response (IRITD3, IRITD5, ENDATs; decreased KTs), coinciding with the emergence of the interstitial infiltrate.

Thus in allografts, alloimmune T cells organized T cell and macrophage interstitial infiltration by day 3, which triggered the active injury–repair response in kidney tissue. The increase in CATs and GRITs reached a plateau after day 5–7, while expression of AMATs continued to increase as tubulitis progressed (23). The coexistence of strong ‘classical’ (IFNG induced) and alternative macrophage activation features, previously considered mutually exclusive, illustrates how immunologic paradigms must be reformulated to accommodate the realities in inflamed organs.

Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References

Microarrays provide a window on the biology operating in biopsies, revealing molecule and PBT changes in relationship to one another and to histopathology, function, and outcomes (24). PBTs correlated strongly with one another and with histopathology (Figure 2). Many injuries and diseases shared this stereotyped disturbance, involving T cell and macrophage transcripts and IFNG induced transcripts, which was continuous across nonrejection, ABMR, and TCMR. The stereotyped active injury–repair response (IRITD3, IRITD5, ENDAT, KT1, KT2) was a prominent feature in human biopsies for clinical indications (BFC) across many disparate diseases and injuries.

image

Figure 2. Relationship between PBT scores and histopathologic diagnosis in human kidney transplant biopsies for cause. Biopsies (n = 234) were sorted by the T cell burden (QCAT score from lowest to highest). Scores are shown for other PBTs in each biopsy (QCAT, GRIT1, QCMAT, AMAT1, IRITD5, IRITD3, ENDAT, KT1 and KT2). As controls, probesets not passing IQR filtering are illustrated for each biopsy. The ribbon above the graph illustrates the relationship of the PBT scores to histopathology diagnosis.

Download figure to PowerPoint

The changes were greatest in TCMR but with no absolute specificity: GRITs and QCATs were also elevated in ABMR and other conditions. This is not because TCMR always coexists with ABMR and other diseases, but because many injuries and diseases induce the active injury–repair response, with secondary inflammation resembling mild TCMR. The histologic feature best correlating with the overall disturbance was the total i-score (25).

Rejection

We distinguished rejection from nonrejection using predictive analysis of microarrays (26). The genes distinguishing rejection from nonrejection were mainly GRITs (e.g. CXCL9, CXCL11, GBP1, INDO), probably because both TCMR and ABMR induce IFNG release: presumably in TCMR from T cells, and in ABMR from NK cells triggered through their Fc receptors.

TCMR:  TCMR is characterized by high expression of T cell, macrophage, and IFNG-inducible transcripts, plus the active injury–repair response in the parenchyma. QCAT expression in human biopsies with TCMR (27) resembled CTL RNA diluted in kidney RNA, never exceeding 15% of the values in pure CTL, and correlating with infiltration. QCAT scores correlated with scores for individual transcripts, for example GZMB. Thus the burden of T cells can be estimated by the mean QCAT score, but also by any one QCAT. NK selective transcripts were poorly expressed in TCMR, consistent with the paucity of NK cells by immunostaining.

We defined the fully developed or canonical form of TCMR exclusively by two molecular features identified in severe TCMR in mouse kidney allografts: intense co-expression of GRITs and AMATs (18). Without knowledge of the histopathology findings, these molecular features identified human biopsies displaying TCMR features, both molecular (e.g. elevated QCATs) and histologic (infiltrate, tubulitis). When cases with ABMR were excluded, canonical TCMR had an excellent prognosis in adherent patients, even if vasculitis was present or if the rejection episodes were late. Thus an exclusively molecular definition of TCMR confirmed the validity of the Banff diagnostic lesions, and underscored the favorable prognosis of TCMR when ABMR and nonadherence are excluded.

ABMR:  This phenotype must be defined in patients because it cannot currently be modeled in mice. Alloantibody acting on the microcirculation in ABMR induced higher expression of ENDATs in ABMR than in other diseases (21). Thus ENDATs were increased in all injured kidneys but were highest in ABMR, correlating with histopathologic lesions of ABMR and with DSA. Many kidneys with high ENDATs and ABMR lesions were C4d negative, indicating that C4d staining is not sensitive for ABMR. Recently we also looked for transcripts associated with donor-specific HLA antibodies and found NK cell transcripts as well as ENDATs (28).

Correlates of atrophy-fibrosis and of time of the biopsy posttransplant

Atrophy-scarring correlated with transcripts associated with B cells, plasma cells, mast cells, as well as other transcripts of unknown significance (29). The strongest correlation was with mast cell transcripts, reflecting the propensity of mast cells to locate in areas of atrophy-fibrosis.

Because of the reported association of B cell infiltration with poor prognosis (8), we studied the significance of B cell associated transcripts (BATs) and immunoglobulin transcripts (IGTs). BATs and IGTs correlated with immunostaining for B cells and plasma cells respectively, and with expression of B cell and plasma cell transcription factors (30). BATs and IGTs were increased in kidneys with atrophy-fibrosis, accumulating in areas of fibrosis, correlating with time posttransplant >5 months and with inflammation in scarred kidneys, but did not correlate with ABMR or prognosis. Similarly, FOXP3 mRNA (31) was a feature of late BFC with scarring and inflammation, but when time was considered did not correlate with outcomes.

Thus the data-driven approach at present indicates that the many inflammatory cell changes such as plasma cells and mast cells that occur in kidneys with extensive nephron loss (atrophy-scarring) may be like the fish inside a sinking ship: they appear late and may be secondary to the destruction, not related to disease.

GFR:  Impaired renal function correlated best with the active injury–repair response, not rejection. Most individual transcripts with strong correlations with function had also been annotated as IRITD3 or IRITD5 (32). Our interpretation is that diseases such as TCMR impair function by inducing a stereotyped injury–repair response in the parenchyma, which we believe is the common pathway to functional disturbances.

Progression to failure

Kidney transplants that require a BFC more than one year posttransplant have a high risk of progression to failure (33). We used supervised principal component analysis to derive a risk score classifier to predict graft loss. The genes most strongly predicting risk of future progression reflected the active injury–repair response, not inflammation (34), including some well known injury-induced molecules such HAVCR1 (KIM1) (33). The risk score in late kidney transplants correlated with atrophy-fibrosis, proteinuria, and GFR, but in multivariable analysis the risk score was the best predictor of outcome, cancelling most variables currently used for predicting outcome, including atrophy-fibrosis and GFR.

The fact that molecules of the injury–repair response were the strongest predictor of future failure after BFC was surprising, because this response was defined in mouse isografts and ATN, where injury is reversible. Our interpretation is that progression in renal diseases, such as late ABMR or recurrent GN occurs via increments of initially reversible nephron damage, manifested by the active injury–repair response. Damage becomes irreversible as nephrons reach their finite limits of repair and shut down irreversibly, leaving atrophy-fibrosis. Risk of failure after a BFC correlates best with ongoing injury–repair response as an indicator of ongoing damage. Thus the active injury–repair response induced by diseases that are treatable (like TCMR) or self limited (like ATN) will resolve, whereas the response induced by persistent stresses such as ABMR is an ominous indicator that the disease is actively damaging the parenchyma.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References

Microarrays validate many features of histopathology, but also reveal an iceberg of changes largely invisible to histopathology (Table 2). Molecular changes are stereotyped, because many types of injury and disease trigger the active injury–repair response in parenchymal, stromal, and microvascular cells, including secondary inflammation. When repair cannot be achieved, nephrons are lost, leading to a cumulative burden of injury (atrophy-scarring), which includes some inflammatory features, such as B cells, plasma cells, and mast cells. Superimposed on these patterns are features that are quantitatively specific for diseases. Rejection is best distinguished from other conditions by intense IFNG effects. TCMR and ABMR are distinguished by greater burden of T cell and macrophage transcripts in TCMR, and the greater endothelial and NK transcripts in ABMR. Function at time of biopsy and the probability of progression after biopsy correlate best with the injury–repair response.

Table 2.  The molecular phenotype of renal transplant BFC: new insights
1. The molecular phenotype of diseased tissue is shaped by three elements:
 a. specific diseases, for example TCMR, ABMR;
 b. active injury–repair response, with secondary inflammation;
 c. cumulative burden of injury, with secondary inflammation.
2. Molecular changes tend to affect large groups of molecules simultaneously: single molecule changes almost always represent changes in large groups of molecules.
3. Rejection (TCMR, ABMR, mixed) is distinguished from other conditions by greater IFNG effects
4. TCMR is distinguished from ABMR by
 a. greater changes in T cell and macrophage molecules in TCMR
 b. greater changes in ENDATs and NK transcripts in ABMR
5. The active injury–repair response is
 a. the strongest correlate with the functional disturbance at the time of BFC
 b. the strongest correlate of progression to failure after late BFC
 c. better assessed by molecular changes than by histopathology
6. Kidneys with extensive scarring manifest transcripts reflecting inflammatory changes (e.g. plasma cell and mast cell infiltration) that appear late and are secondary to nephron destruction, like fish in a sunken ship, and should not be confused with the cause, for example ABMR.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References

Thanks to Jessica Chang for preparing the figures, and to Danielle Stewart for preparing the manuscript. This work was supported by funding and/or resources from Genome Canada, Genome Alberta, the University of Alberta, the University of Alberta Hospital Foundation, Roche Molecular Systems, Hoffmann-La Roche Canada Ltd., Alberta Ministry of Advanced Education and Technology, the Roche Organ Transplantation Research Foundation, the Kidney Foundation of Canada, Stromedix, and Astellas. Dr. Halloran also holds the Muttart Chair in Clinical Immunology.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References

The authors of this manuscript have conflicts of interest to disclose as described by the American Journal of Transplantation. P.F.H. is the chief executive officer of Transcriptome Sciences Inc, a company with an interest in molecular diagnostics.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Deconstructing Biologic Events in Disease: Pathogenesis-Based Transcript Sets (PBTs)
  5. Summarizing the PBT Changes in Mouse Kidney Transplants
  6. Molecular Features of Human Kidney Transplant Biopsies for Clinical Indications
  7. Conclusions
  8. Acknowledgments
  9. Disclosure
  10. References